A material recommendation method and device, computing device, and storage medium

Abstract

Embodiments of the present application disclose a material recommendation method and device, a computing device and a storage medium. The method comprises: in response to an operation of a target user entering a target page, screening materials meeting a first preset condition from a resource pool as first materials; screening materials meeting a second preset condition from an invalid resource pool as second materials; screening materials associated with the target user from the first materials and the second materials as third materials; and sorting the third materials according to material quality, and recommending the first target number of sorted materials to the target user. It can be known that the high-quality materials removed from the recommendation system are re-screened, combined with the materials in the material pool, and the materials associated with the user are selected from the combined materials to be recommended to the user. The high-quality materials removed from the recommendation system can be recommended to the user again, the quality of the materials recommended to the user is improved, the user can quickly locate the interested content, and the user stickiness is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, and in particular, to a material recommendation method and device, a computing device and a storage medium. BACKGROUND

[0002] Material refers to the content uploaded to the network platform by users and creators, including articles, videos, etc. The material can be viewed by other users to understand relevant information. Excellent material can be recommended to users to cause emotional resonance of users, thereby improving the stickiness of users to the network platform.

[0003] The existing material recommendation process includes: in response to an event that a target user enters a target page, screening materials that do not contain sensitive fields such as politics and pornography from a resource pool storing materials within a preset time period, and further screening materials related to the preferences of the target user from the screened materials through a recall algorithm. Then, the materials screened twice are sorted according to the quality of the materials, and the top target number of materials after sorting are recommended to the target user. However, in the above material recommendation process, materials will be removed over time, which will cause some excellent materials to be removed over time, so that they will not be recommended to users, reducing the quality of the materials recommended to users, causing users to be unable to quickly find interesting content, easily causing users to lose, and reducing user stickiness.

[0004] Therefore, it is desirable to provide a material recommendation method to solve the technical problem of reducing user stickiness in the existing material recommendation method. SUMMARY

[0005] To this end, the embodiments of the present application provide a material recommendation method, device, computing device and storage medium to try to solve or at least alleviate the above problems.

[0006] According to an aspect of the embodiments of the present application, a material recommendation method is provided, which is adapted to be executed in a computing device, and the computing device stores a resource pool and an invalid material pool. The resource pool stores materials generated within a first preset time, and the invalid material pool stores materials not recommended within a second preset time. The method comprises: in response to an operation of a target user entering a target page, screening materials meeting a first preset condition from the resource pool as first materials; screening materials meeting a second preset condition from the invalid resource pool as second materials; screening materials associated with the target user from the first materials and the second materials as third materials; and sorting the third materials according to the quality of the materials, and recommending the top target number of third materials after sorting to the target user.

[0007] Optionally, the first preset condition comprises that the sensitive field is not included in the material, and the second preset condition comprises that an index value of the material exceeds an index threshold corresponding to the material, the index value being calculated according to a preset evaluation index and a decay coefficient, the decay coefficient being negatively correlated with a generation time length of the material.

[0008] Optionally, the screening of the material meeting the second preset condition from the invalid resource pool comprises: determining an index value and an index threshold of a first target material in the invalid resource pool; determining whether the first target material meets the second preset condition, and if so, marking the first target material; after the marking of the first target material, or if the first target material does not meet the second preset condition, determining a new first target material, and determining an index value and an index threshold of the new first target material, until all materials in the invalid resource pool are traversed; and screening the marked material from the invalid resource pool as the second material.

[0009] Optionally, the preset evaluation index comprises a material click-through rate index, a per-viewer viewing time length index, a manual evaluation score index, and a comment interaction behavior index.

[0010] Optionally, the calculation formula of the index value comprises: ω = (α * x + β * y + δ * z + ε * w) * μ, wherein ω represents, α, β, δ, and ε represent, x represents a value of the material click-through rate index, y represents a value of the per-viewer viewing time length index, z represents a value of the manual evaluation score index, w represents a value of the comment interaction behavior index, and μ represents a decay coefficient, the decay coefficient being an exponential function with the generation time length of the material as an index and with any value greater than 0 and less than 1 as a coefficient.

[0011] Optionally, the computing device further stores a material type and index threshold correspondence relationship, and the determination of the index threshold of the first target material in the invalid resource pool comprises: determining a material type and a generation time length of the first target material; and searching, from the material type and index threshold correspondence relationship, for an index threshold corresponding to the determined material type and generation time length.

[0012] Optionally, the sorting of the third material according to the material quality and the recommendation of the first target number of third materials in the sorted order to the target user comprises: determining the quality of the third material according to a value of any one or more indexes in the preset evaluation index of the third material; sorting the third material according to the material quality from high to low; and recommending the first target number of third materials in the sorted order to the target user.

[0013] Optionally, the material recommendation method further comprises: screening the marked material from the materials recommended to the target user as a second target material; calculating the index value of the second target material; judging whether the second target material meets the second preset condition, and if not, deleting the second target material from the materials recommended to the target user.

[0014] According to another aspect of the present application, there is provided a material recommendation device adapted to be executed in a computing device, the computing device storing a resource pool and a failed material pool, the resource pool storing materials generated within a first preset time, and the failed material pool storing materials not recommended within a second preset time, the device comprising: a legal material determination module adapted to screen materials meeting a first preset condition from the resource pool as first materials in response to an operation of a target user entering a target page; a material backtracking module adapted to screen materials meeting a second preset condition from the failed resource pool as second materials; a material to be recommended determination module adapted to screen materials associated with the target user from the first materials and the second materials as third materials; and a material recommendation module adapted to sort the third materials according to material quality and recommend the top target number of the sorted third materials to the target user.

[0015] According to another aspect of the present application, there is provided a computing device comprising: at least one processor; and a memory storing program instructions, wherein the program instructions are configured to be executed by the at least one processor, and the program instructions comprise instructions for executing the method as described above.

[0016] According to another aspect of the present application, there is provided a readable storage medium storing program instructions, which, when read and executed by a computing device, cause the computing device to execute the method as described above.

[0017] According to the material recommendation method of the present application, first, in response to an operation of a target user entering a target page, materials meeting a first preset condition are screened from a resource pool as first materials, then materials meeting a second preset condition are screened from a failed resource pool as second materials, and materials associated with the target user are screened from the first materials and the second materials as third materials, finally, the third materials are sorted according to material quality, and the top target number of the sorted third materials are recommended to the target user.

[0018] As can be seen from the above, the present application reselects high-quality materials removed from a recommendation system, combines the reselected materials with materials in a material pool, and screens materials associated with a user from the combined materials to recommend to the user, so that the high-quality materials removed from the recommendation system can be recommended to the user again, the quality of the materials recommended to the user is improved, the user can quickly find interested content, and user stickiness is improved.

[0019] Secondly, the application carries out backtracking on the materials removed from the resource pool according to the material click-through rate index, the per capita viewing time index, the artificial evaluation score index and the comment interaction behavior index, which is equivalent to backtracking on the materials removed from the resource pool from multiple dimensions, and guarantees the quality of the backtracked materials.

[0020] In addition, the backtracked materials are re-screened in the application, and the backtracked materials with poor performance are deleted, so that more high-quality materials can be recommended for the target user, the quality of the materials recommended for the user is improved, and the user stickiness is further improved.

[0021] The above description is only a summary of the technical scheme of the embodiments of the application, in order to more clearly understand the technical means of the embodiments of the application, the content of the specification can be implemented, and in order to make the above and other purposes, characteristics and advantages of the embodiments of the application more obvious and easy to understand, the specific implementation manner of the embodiments of the application is described below. BRIEF DESCRIPTION OF DRAWINGS

[0022] In order to achieve the above and related purposes, certain illustrative aspects will be described herein in connection with the following description and drawings, which indicate various ways in which the principles disclosed herein can be practiced, and all aspects and their equivalents are intended to fall within the scope of the claimed subject matter. The above and other objects, features and advantages of the present disclosure will become more apparent from the following detailed description taken in conjunction with the accompanying drawings. Throughout the disclosure, the same reference numerals are generally used to refer to similar or like components throughout.

[0023] Figure 1 A block diagram of a computing device 100 according to one embodiment of the application is shown;

[0024] Figure 2 A flowchart of a material recommendation method 200 according to one embodiment of the application is shown;

[0025] Figure 3 A flowchart of screening materials meeting a second preset condition from a failed resource pool according to one embodiment of the application is shown;

[0026] Figure 4 A structural block diagram of a material recommendation device 400 according to one embodiment of the application is shown. DETAILED DESCRIPTION

[0027] Exemplary embodiments of the present disclosure will be described below in greater detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it is to be understood that the present disclosure can be embodied in various forms without being limited by the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure will be thorough and complete, and will fully convey the scope of the present disclosure to those skilled in the art.

[0028] In the present disclosure, in order to solve the technical problem of reducing user stickiness of the existing material recommendation method, a material recommendation method is provided. The method can be implemented as an application installed in a computing device 100 and jointly complete network communication line checking with a target application. The method can also be implemented as a plug-in integrated in a target application program running on the computing device 100. Here, the target application program can be understood as an application program including a material recommendation function, and the present disclosure does not limit the target application program. For example, the target application program can be a car home application program.

[0029] The starting mode of the material recommendation method can be set according to the actual application scenario, and the present disclosure does not limit this. For example, when the material recommendation method is implemented as an application, the method is automatically started when the computing device is started, i.e., the computing device is powered on and successfully logs in the user and enters the user interface. When the material recommendation method is implemented as a plug-in, the method is automatically started when the target application program in the computing device is started.

[0030] The computing device 100 described above can be implemented as a server, such as an application server, a Web server, etc.; or as a desktop computer, a notebook computer, a processor chip, a tablet computer, etc., but is not limited thereto. Figure 1 A block diagram of the physical components (i.e., hardware) of the computing device 100 is shown. In a basic configuration, the computing device 100 includes at least one processing unit 102 and a system memory 104. According to an aspect, depending on the configuration and type of computing device, the system memory 104 includes, but is not limited to, volatile storage (e.g., random access memory), non-volatile storage (e.g., read-only memory), flash memory, or any combination of such memories.

[0031] According to an aspect, the system memory 104 includes an operating system 105. The system memory 204 also includes an application 150. According to an aspect, the operating system 105, for example, is suitable for controlling the operation of the computing device 100. Moreover, the example is practiced in conjunction with a graphics library, other operating systems, or any other application program, and is not limited to any particular application or system. In Figure 1The basic configuration is shown in FIG. 1 by those components within the dashed line 108. According to one aspect, the computing device 100 has additional features or functionality. For example, according to one aspect, the computing device 100 includes additional data storage devices (removable and / or non-removable) such as, for example, magnetic disks, optical disks, or tape. Such additional storage is illustrated in FIG. 1 by the removable storage 109 and the non-removable storage 110. Figure 1

[0032] As stated above, according to one aspect, a plurality of program modules is stored in the system memory 104. When the identity authentication method provided by the present application is implemented as an application program 150, and is executed by the processing unit 102, the application program 150 performs a process, which includes but is not limited to one or more stages of the method 200. According to one aspect, the type of application program is not limited, for example, the application program also includes: email and contact application program, word processing application program, spreadsheet application program, database application program, slide show application program, drawing or computer-aided application program, web browser application program, etc.

[0033] According to one aspect, the examples can be practiced with circuitry that includes discrete electronic elements, a Figure 1 According to one aspect, the examples can be practiced with circuitry that includes discrete electronic elements, a

[0034] ​According to an aspect, the computing device 100 can also have one or more input device(s) 112 such as a keyboard, a mouse, a pen, a voice input device, a touch input device, etc. Notably, the touch input device can be a touch screen, and the touch screen in the present application can be divided into multiple areas, and authentication information can be input in each area. Output device(s) 114 such as a display, speakers, a printer, etc. can also be included. The aforementioned devices are examples and others can be used. The computing device 100 can include one or more communication connections 116 allowing communication with other computing devices 118, examples of which include, but are not limited to: RF transmitter, receiver, and / or transceiver circuitry; universal serial bus (USB), parallel, and / or serial ports. The computing device 100 can be connected with one or more other computing devices 118 in communication therewith.

[0035] The term computer readable media as used herein includes computer storage media. Computer storage media can include volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, or program modules. The system memory 104, the removable storage device 109, and the non-removable storage device 110 are all computer storage media examples (i.e., memory storage.) Computer storage media can include Random Access Memory (RAM), Read Only Memory (ROM), Electronically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store information and which can be accessed by computing device 100. According to an aspect, any such computer storage media can be part of the computing device 100. Computer storage media does not include a carrier wave or other propagated data signal.

[0036] According to an aspect, communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media. According to an aspect, the term "modulated data signal" describes a signal that has one or more characteristics set or changed in such a manner as to encode information in the signal. By way of example, and not limitation, communication media includes wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, radio frequency (RF), infrared, and other wireless media.

[0037] It should be noted that the resource pool and the invalid material pool are stored in the computing device or a database in communication connection with the computing device. Since the material with a long interval creation time will lose validity, the material in the resource pool is the material created by the user within a first preset time. The first preset time can be set according to an actual application scenario, and the present application does not limit this. For example, the first preset time can be 6 months.

[0038] The invalid material pool can be understood as storing the material removed from the recommendation system, which is equivalent to storing the material not recommended to the user. Similarly, since the material with a long interval creation time will lose validity, the material in the invalid material pool is the material not recommended to the user within a second preset time, so that the number of invalid materials can be well controlled. The second preset time can be set according to an actual application scenario, and the present application does not limit this. For example, the second preset time can be 60 days.

[0039] The above-mentioned resource pool and invalid material pool can be understood as storage containers, which can be cache blocks or memory blocks, and of course, can also be implemented in other forms, and the present application is not limited thereto.

[0040] The computing device or the database in communication connection with the computing device also stores a material type and index threshold value correspondence relationship, which includes the fields: material type, material generation time period and index threshold value. The material type can be divided according to an actual application scenario, and the present application does not limit this. For example, the material type includes price reduction promotion material, vehicle evaluation material, etc. The material generation time period refers to the time period of the material creation completion time, specifically the number of days of the material creation completion, for example, 30 days to 60 days belong to a material generation time period. The index threshold value is the basis for screening the material from the invalid resource pool to be recommended to the user again. Only when the index value of the material exceeds the index threshold value, the material removed from the resource pool will be selected again, that is, only when the index value of the material exceeds the index threshold value, the material will be traced back. The calculation method of the index value of the material will be described below. The specific existence form of the material type and index threshold value correspondence relationship can be set according to an actual application scenario, and the present application does not limit this. For example, the material type and index threshold value correspondence relationship can be implemented as a data table. Part of the content of the material type and index threshold value correspondence table is shown in Table 1.

[0041] Table 1

[0042] Material Type Material Generation Time Period / Days Indicator Threshold Downscale Promotion Material Less than 30 1.5 30-60 2.0 Greater than 60 2.5 Vehicle Evaluation Material Less than 180 1.6 180-365 2.0 Greater than 365 2.4

[0043] As shown in Table 1, the material type and index threshold value correspondence table provided by the embodiment of the present application, for different material types, multiple material generation time periods are divided, and for different types of materials, different index threshold values are corresponded in different material generation time periods. Specifically, for the reduced price promotion material, as the material generation time increases, the corresponding index threshold value increases, which is equivalent to increasing the requirement for material re-selection, and then the opportunity for re-selection is correspondingly reduced.

[0044] The index value of the material is calculated according to the preset evaluation index and the decay coefficient. The preset evaluation index includes the material click-through rate index, the average viewing time per person index, the artificial evaluation score index, and the comment interaction behavior index. The decay coefficient is an exponential function with the material generation time as the index and any value in the interval greater than 0 and less than 1 as the coefficient, for example, the coefficient of the exponential function is 0.7. The calculation formula of the index value of the material includes:

[0045] ω=(α*x+β*y+δ*z+ε*w)*μ

[0046] In the formula, ω represents, α, β, δ, ε respectively represent the coefficients of the material click-through rate index, the average viewing time per person index, the artificial evaluation score index, and the comment interaction behavior index. The coefficients of each index can be set according to the actual application scene and the degree of influence of each index on the quality of the material, for example, α, β, δ, ε are 1.8, 1.7, 1.6, and 1.5, of course, the present application is not limited to this. x represents the value of the material click-through rate index, y represents the value of the average viewing time per person index, z represents the value of the artificial evaluation score index, w represents the value of the comment interaction behavior index, and μ represents the decay coefficient.

[0047] Then, according to the above formula, as the material generation time increases, the decay coefficient decreases, so that as the material generation time increases, the index value of the material decreases, and as the material generation time increases, the index threshold value corresponding to the material increases, so that the possibility of the index value of the material exceeding the index threshold value decreases, and therefore, as the material generation time increases, the possibility of the material being traced back is lower, which also conforms to the timeliness of the material. The timeliness of the material can be understood as the time when the material has the maximum performance value. That is, the material after the timeliness is difficult to be traced back, so as to achieve the purpose of screening the optimal material from the invalid resource pool.

[0048] Figure 2 A flowchart of a material recommendation method 200 according to an embodiment of the present application is shown. The method 200 is suitable for being executed in the computing device 100. The method 200 can include steps 210 to 240.

[0049] The material recommendation method provided by the application reselects high-quality materials that have been removed from the recommendation system, combines the materials with materials in a material pool, and selects materials associated with a user from the combined materials to recommend to the user, so that the high-quality materials removed from the recommendation system can be recommended to the user again, the quality of the materials recommended to the user is improved, the user can quickly find interesting content, and user stickiness is improved.

[0050] When the target user enters the target page, in response to the operation, step 210 is performed to select a material meeting a first preset condition from the resource pool as a first material. The first preset condition includes that the material does not contain a sensitive field, that is, step 210 is to select a material not containing a sensitive field from the resource pool. The sensitive field can include a political-related field and a yellow-related field.

[0051] After the material without the sensitive field is selected, step 220 is performed to select a material meeting a second preset condition from the invalid resource pool as a second material. The second preset condition includes that an index value of the material exceeds an index threshold corresponding to the material. The calculation formula of the index value is described above, and will not be described here.

[0052] Figure 3 A flowchart of selecting a material meeting a second preset condition from an invalid resource pool according to an embodiment of the application is shown. Figure 3 Steps 310 to 370 are included.

[0053] First, step 310 is performed to determine a material from the invalid resource pool as a first target material.

[0054] After the first target material in the invalid material pool is determined, step 320 is performed to determine a material type and a material generation duration of the first target material. That is, the material type and the material generation duration of any material in the invalid material pool are determined.

[0055] Then, step 330 is performed to find an index threshold corresponding to the determined material type and the material generation duration from a material type and index threshold correspondence. After the index threshold is determined, the value of the preset evaluation index of the first target material is determined, and the index value of the first target material is calculated according to the calculation formula ω=(α*x+β*y+δ*z+ε*w)*μ.

[0056] After the index value of the first target material and the index threshold are determined, step 340 is performed to determine whether the index value of the first target material exceeds the index threshold corresponding to the first target material, that is, whether the first target material meets the second preset condition.

[0057] If the index value of the first target material exceeds the index threshold corresponding thereto, i.e., the first target material meets the second preset condition, step 350 is performed to mark the first target material. The specific implementation manner of marking can be set according to actual application scenarios, and the present application does not limit this. For example, a preset field is added to the material to achieve the purpose of marking.

[0058] After the first target material is marked, or if the index value of the first target material is less than the index threshold corresponding thereto, i.e., the first target material does not meet the second preset condition, step 360 is continued to determine whether all materials in the invalid resource pool have been traversed. If yes, step 370 is performed to filter out the marked materials from the invalid resource pool as second materials. Then the execution of the program is ended. If all materials in the invalid resource pool have not been traversed, step 310 is continued to determine a material from the invalid resource pool as the first target material.

[0059] Thus, steps 310 to 370 are used to filter out materials meeting the second preset condition from the invalid resource pool.

[0060] From the above content, it can be seen that in the embodiment of the present application, the materials removed from the resource pool are traced back according to the material click-through rate index, the per capita viewing time index, the artificial evaluation score index and the comment interaction behavior index, which is equivalent to tracing back the materials removed from the resource pool from multiple dimensions, thereby ensuring the quality of the traced back materials. The material tracing back can be understood as reselecting the materials removed from the resource pool to the resource pool.

[0061] After the second materials meeting the second preset condition are filtered out from the invalid resource pool, step 230 is continued to filter out the materials associated with the target user from the first materials and the second materials as third materials. That is, the first materials and the second materials are combined, and then the materials associated with the target user are filtered out therefrom.

[0062] In some embodiments, step 230 includes filtering out the materials associated with the target user from the first materials and the second materials by a multi-recall algorithm, and the filtered out materials are the third materials. The recall algorithm can include a single Embedding vector recall algorithm, a multi Embedding vector recall algorithm and a TDM deep tree matching recall algorithm, etc. The filtering out of the materials associated with the target user from the first materials and the second materials by the recall algorithm is prior art, which will not be described here again, but is within the protection scope of the present application.

[0063] After the third material associated with the target user is screened out, step 240 is performed to sort the third materials according to the material quality, and the top target number of the sorted third materials are recommended to the target user. The target number can be set according to the actual application scenario, and the present application does not limit this. For example, the target number is 10.

[0064] In some embodiments, step 240 includes determining the quality of the third material according to the value of any one or more of the preset evaluation indicators of the third material, including the material click-through rate indicator, the per capita viewing time indicator, the artificial evaluation score indicator, and the comment interaction behavior indicator. For example, if the material click-through rate indicator is used to determine the material quality, the material click-through rate of each third material can be used as the material quality value, and the top target number of the sorted third materials are recommended to the target user according to the order from high to low. Of course, the sum of multiple indicator values can also be used as the quality value of the material.

[0065] In order to recommend high-quality materials to the target user, in some embodiments, after the material is traced back, the traced back material recommended to the user can be understood as an exposed material, and then the exposed material is re-screened. Specifically, first, the marked material is screened out from the materials recommended to the target user as the second target material, that is, the traced back material is screened out from the materials recommended to the user. Then, the index value of each second target material is calculated according to the calculation formula ω=(α*x+β*y+δ*z+ε*w)*μ, and it is judged whether the index value of each second target material exceeds the corresponding index threshold value. If it does not exceed, it means that the second target material (i.e. the traced back material) performs poorly, and then the second target material is deleted from the materials recommended to the target user. If it exceeds, it means that the traced back material performs well, and then the traced back material does not need to be deleted, and the execution of the program is ended.

[0066] From the above, it can be seen that the traced back material is re-screened in the present application, and the traced back material that performs poorly is deleted, so that more high-quality materials can be recommended to the user according to the real-time behavior of the user, the quality of the materials recommended to the user is improved, and the user stickiness is further improved.

[0067] At this point, the material recommendation is realized. Whenever material recommendation is needed, steps 210 to 240 are executed.

[0068] Figure 4 The structural block diagram of a material recommendation device 400 according to an embodiment of the present application is shown. The device 400 includes a legal material determination module 410, a material tracing back module 420, a to-be-recommended material determination module 430, and a material recommendation module 440 coupled in sequence.

[0069] The legal material determining module 410 is suitable for screening materials meeting a first preset condition from the resource pool as first materials in response to an operation of the target user entering the target page.

[0070] The material backtracking module 420 is suitable for screening materials meeting a second preset condition from the invalid resource pool as second materials.

[0071] The material to be recommended determining module 430 screens materials associated with the target user from the first materials and the second materials as third materials.

[0072] The material recommendation module 440 is suitable for sorting the third materials according to material quality and recommending the top target number of the sorted materials to the target user.

[0073] It should be noted that the working principle of the material recommendation device 400 is similar to the above-mentioned material recommendation method 200, and the relevant parts can refer to the description of the above-mentioned material recommendation method 200, which will not be repeated here.

[0074] From the above, the material recommendation method provided by the application reselects high-quality materials that have been removed from the recommendation system, combines the materials in the material pool, and then selects materials associated with the user from the combined materials and recommends them to the user, so that the high-quality materials removed from the recommendation system can be recommended to the user again, improving the quality of the materials recommended to the user, facilitating the user to quickly find interesting content, and improving user stickiness.

[0075] Secondly, the application backtracks the materials removed from the resource pool according to the material click-through rate index, the per capita viewing time index, the artificial evaluation score index and the comment interaction behavior index, which is equivalent to backtracking the materials removed from the resource pool from multiple dimensions, ensuring the quality of the backtracked materials.

[0076] Moreover, the backtracked materials are reselected in the application, and the backtracked materials that perform poorly are deleted, so that more high-quality materials can be recommended to the target user, improving the quality of the materials recommended to the user and further improving user stickiness.

[0077] A8 The method of A3 further comprises:

[0078] Screening the marked materials from the materials recommended to the target user as second target materials;

[0079] Calculating the index value of the second target material;

[0080] Determining whether the second target material meets the second preset condition, and if not, deleting the second target material from the materials recommended to the target user.

[0081] The various techniques described herein can be implemented in connection with hardware or software or, where appropriate, with a combination of both. Thus, the methods and apparatus of the application, or certain aspects or portions thereof, can take the form of program code (i.e., instructions) embodied in tangible media, such as removable hard disks, USB flash drives, optical tapes, CD-ROMs, or any other machine-readable storage medium, wherein, when the program code is loaded into and executed by a machine, such as a computer, the machine becomes an apparatus for practicing the subject application.

[0082] Where the program code is executed on a programmable computer, the computing device generally includes a processor, a storage medium readable by the processor (including volatile and non-volatile memory and / or storage elements), at least one input device, and at least one output device. The storage medium is configured to store program code which, when executed by the processor, carries out methods of the subject application.

[0083] In a storage medium readable by a machine, the program code can be implemented using any of a number of high level, low level, object-oriented, updated structured, or other programming languages. Examples of possible programming languages include C, C++, C#, Java, Visual Basic, Python, SQL, and JavaScript. Program code for carrying out operations of the subject application can be written in any combination of one or more programming languages.

[0084] In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the application can be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been described in detail in order to avoid obscuring the understanding of this description.

[0085] In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the application can be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been described in detail in order to avoid obscuring the understanding of this description.

[0086] Similarly, it is to be understood that the embodiments of the present application can be alternately grouped into individual embodiments, figures, or descriptions thereof, in order to streamline the disclosure and aid in the understanding of one or more of the individual inventive aspects. Nevertheless, the method of this disclosure is not to be interpreted to reflect an intention that the claimed application requires more features than are recited in each of the claims.

[0087] Those skilled in the art understand that the modules, or units, or components of the devices in the examples disclosed herein can be arranged in the devices as described in the examples, or alternatively can be located in one or more devices different from the devices in the examples. The modules in the foregoing examples can be combined into one module or further divided into multiple sub-modules.

[0088] Those skilled in the art understand that the modules in the devices in the examples can be adaptively changed and arranged in one or more devices different from the examples. The modules or units or components in the examples can be combined into one module or unit or component, and further divided into multiple sub-modules or sub-units or sub-components. All the features disclosed in this specification (including the claims, abstract, and drawings) and all the processes or units of any method or device so disclosed can be combined in any combination, excepting that at least some of such features and / or processes or units are mutually exclusive. Each feature disclosed in this specification (including the claims, abstract, and drawings) can be replaced by alternative features providing the same, equivalent, or similar functionality unless expressly stated otherwise.

[0089] Further, those skilled in the art understand that the combination of features of different embodiments means within the scope of the present application and forms different embodiments, although some of the examples described herein include certain features rather than others included in other examples. For example, in the claims, any one of the claimed embodiments can be used in any combination.

[0090] Further, some of the examples described herein are described as combinations of methods or method elements implemented by a processor of a computer system or by other means for performing the functions described. Accordingly, a processor with the necessary instructions for performing such methods or method elements forms a means for performing the methods or method elements.

[0091] As used herein, unless otherwise indicated, the use of the ordinal adjectives "first", "second", "third", etc., are to add specificity and difference to a term, and are not intended to indicate a temporal or chronological order or sequence. Except to the extent necessary or implicit from the discussion herein, as readily apparent to those of ordinary skill in the art, no portion of the specification is indicative of an ordering.

[0092] While the present application has been described in connection with limited number of embodiments, those skilled in the art will appreciate that numerous modifications and variations therefrom can be made without departing from the scope of the present application as set forth in the limitations to follow. Moreover, the language used in this specification has been chosen for readability and instructional purposes and can not have been selected to delineate or circumscribe the subject application. Accordingly, the disclosure of the present application is intended to be illustrative, but not limiting, of the scope of the application, which is set forth with particularity in the appended claims.

Claims

1. A material recommendation method suitable for being executed in a computing device, the computing device storing a resource pool and a failed resource pool, the resource pool storing materials generated within a first preset time, the failed resource pool storing materials not recommended within a second preset time, the method comprising: in response to an operation of a target user entering a target page, screening materials meeting a first preset condition from the resource pool as first materials; screening materials meeting a second preset condition from the failed resource pool as second materials; screening materials associated with the target user from the first materials and the second materials as third materials; sorting the third materials according to material quality, and recommending the first target number of third materials after sorting to the target user; wherein the first preset condition comprises that a material does not contain a sensitive field, the second preset condition comprises that an index value of a material exceeds an index threshold corresponding to the material, the index value is calculated according to a preset evaluation index and a decay coefficient, and the decay coefficient is negatively correlated with a material generation time length. The screening of the materials meeting the second preset condition from the failed resource pool comprises: determining an index value and an index threshold of a first target material in the failed resource pool; determining whether the first target material meets the second preset condition, and if so, marking the first target material; after the marking of the first target material, or if the first target material does not meet the second preset condition, determining a new first target material, and determining an index value and an index threshold of the new first target material until all materials in the failed resource pool are traversed; and screening the marked materials from the failed resource pool as the second materials. The preset evaluation index comprises a material click-through rate index, a per capita viewing time index, a manual evaluation score index, and a comment interaction behavior index. The calculation formula of the index value comprises: ω = (α * x + β * y + δ * z + ε * w) * μ, wherein ω represents the index value, α represents a coefficient of the material click-through rate index, β represents a coefficient of the per capita viewing time index, δ represents a coefficient of the manual evaluation score index, ε represents a coefficient of the comment interaction behavior index, x represents a value of the material click-through rate index, y represents a value of the per capita viewing time index, z represents a value of the manual evaluation score index, w represents a value of the comment interaction behavior index, and μ represents a decay coefficient, the decay coefficient being an exponential function with the material generation time length as an index and any value greater than 0 and less than 1 as a coefficient. The computing device further stores a material type and index threshold correspondence relationship, and the determination of the index threshold of the first target material in the failed resource pool comprises: determining a material type and a material generation time length of the first target material; and searching for an index threshold corresponding to the determined material type and material generation time length from the material type and index threshold correspondence relationship. The sorting of the third materials according to material quality and the recommendation of the first target number of third materials after sorting to the target user comprise:

2. The method of claim 1, wherein, ​ ​ ​ ​ ​ 3. The method of claim 1 or 2, wherein, ​ 4. The method of claim 3, wherein, ​ ​ ​ 5. The method of claim 2, wherein, ​ ​ ​ 6. The method of claim 1, wherein, ​ determine the quality of the third material according to a value of any one or more of the preset evaluation indexes of the third material; sort the third material in descending order of material quality; recommend the top target number of third materials to the target user.

7. The method of claim 2, further comprising: selecting a marked material from the materials recommended to the target user as a second target material; calculating the index value of the second target material; determining whether the second target material meets the second preset condition, and if not, deleting the second target material from the materials recommended to the target user.

8. A material recommendation device adapted to be executed in a computing device, the computing device storing a resource pool and a failed resource pool, the resource pool storing materials generated within a first preset time, and the failed resource pool storing materials not recommended within a second preset time, the device comprising: a legal material determination module adapted to select materials meeting a first preset condition from the resource pool as first materials in response to an operation of a target user entering a target page; a material backtracking module adapted to select materials meeting a second preset condition from the failed resource pool as second materials; a material to be recommended determination module adapted to select materials associated with the target user from the first materials and the second materials as third materials; a material recommendation module adapted to sort the third materials according to material quality and recommend the top target number of sorted materials to the target user; wherein the first preset condition includes that the material does not contain sensitive fields, the second preset condition includes that an index value of the material exceeds a corresponding index threshold, the index value is calculated according to a preset evaluation index and a decay coefficient, and the decay coefficient is negatively correlated with the time length of the material generation.

9. A computing device comprising: at least one processor; and a memory storing program instructions configured to be executed by the at least one processor, the program instructions comprising instructions for executing the method of any one of claims 1 to 7.

10. A readable storage medium storing program instructions, which, when read and executed by a computing device, cause the computing device to execute the method of any one of claims 1 to 7. ​

Images